Traffic event reproduction system, server, traffic event reproduction method, and non-transitory computer readable medium

Abstract

A traffic event reproduction system according to the present procedure includes: a server; a first transmission apparatus configured to transmit first event information including a first event video to the server; a second transmission apparatus configured to transmit second event information including a second event video to the server. A server includes: a reception unit configured to receive first and second event information; a model generation unit configured to generate first and second three-dimensional models including a party to a traffic event extracted from the first and second event videos as an object; a combining unit configured to combine the first and second three-dimensional models to generate a combined three-dimensional model in a three-dimensional space; and a display unit configured to display a video viewed from a user-specified viewpoint using the combined three-dimensional model.

What is claimed is: 
     
       1. A traffic event reproduction system comprising:
 a server; 
 a first transmission apparatus configured to transmit first event information including a first event video of a traffic event to the server; and 
 a second transmission apparatus configured to transmit second event information including a second event video of the traffic event to the server, 
 the server including: 
 at least one memory storing instructions, and 
 at least one processor configured to execute the instructions to: 
 receive the first event information and the second event information from the first transmission apparatus and the second transmission apparatus; 
 generate a first three-dimensional model including an object of a party to the traffic event extracted from the first event video and a second three-dimensional model including an object of the party extracted from the second event video; 
 combine the first three-dimensional model and the second three-dimensional model to generate a combined three-dimensional model in a three-dimensional space including a point corresponding to the occurrence point of the traffic event; and 
 display a video of the party viewed from a user-specified viewpoint on a display using the combined three-dimensional model. 
 
     
     
       2. The traffic event reproduction system according to  claim 1 , wherein
 the first event information further includes first position information indicating the shooting position of the first event video, and the second event information further includes second position information indicating the shooting position of the second event video, and 
 the at least one processor is further configured to execute the instructions to combine the first three-dimensional model and the second three-dimensional model using the first position information and the second position information. 
 
     
     
       3. The traffic event reproduction system according to  claim 2 , wherein
 in combining the first three-dimensional model and the second three-dimensional model, the at least one processor is further configured to execute the instructions to 
 specify the position of the first three-dimensional model in the three-dimensional space on a basis of the first position information and the relative relationship between the shooting position indicated by the first position information and the party extracted from the first event video, and 
 specify the position of the second three-dimensional model in the three-dimensional space on a basis of the second position information and the relative relationship between the shooting position indicated by the second position information and the party extracted from the second event video. 
 
     
     
       4. The traffic event reproduction system according to  claim 2 , wherein
 the first event information further includes first direction information indicating the shooting direction at the shooting position of the first event image, and 
 the second event information further includes second direction information indicating the shooting direction at the shooting position of the second event video, and 
 in combining the first three-dimensional model and the second three-dimensional model, the at least one processor is further configured to execute the instructions to 
 combine the first three-dimensional model and the second three-dimensional model using the first direction information and the second direction information. 
 
     
     
       5. The traffic event reproduction system according to  claim 1 , wherein at least one of the first and second event videos further includes a three-dimensional point cloud image. 
     
     
       6. The traffic event reproduction system according to  claim 1 , wherein at least one of the first and second event videos further includes a two-dimensional camera image and a three-dimensional point cloud image. 
     
     
       7. The traffic event reproduction system according to  claim 6 , wherein the at least one processor is further configured to execute the instructions to extract a texture of the party from the two-dimensional camera image, generate an object of the party by mapping the extracted texture onto the three-dimensional point cloud image, and generate the first three-dimensional model and/or the second three-dimensional model including the generated object. 
     
     
       8. The traffic event reproduction system according to  claim 1 , wherein the first and second transmission apparatuses are respectively installed in at least one of a vehicle of the party or a vehicle in the vicinity of the party. 
     
     
       9. A server comprising:
 at least one memory storing instructions, and 
 at least one processor configured to execute the instructions to: 
 receive first event information including a first event video of a traffic event from a first transmission apparatus, and receive second event information including a second event video of the traffic event from a second transmission apparatus; 
 generate a first three-dimensional model including an object of a party to the traffic event extracted from the first event video and a second three-dimensional model including an object of the party extracted from the second event video; 
 combine the first three-dimensional model and the second three-dimensional model to generate a combined three-dimensional model in a three-dimensional space including a point corresponding to the occurrence point of the traffic event; and 
 display a video of the party viewed from a user-specified viewpoint on a display using the combined three-dimensional model. 
 
     
     
       10. The server according to  claim 9 , wherein
 the first event information further includes first position information indicating the shooting position of the first event video, and the second event information further includes second position information indicating the shooting position of the second event video, and 
 the at least one processor is further configured to execute the instructions to combine the first three-dimensional model and the second three-dimensional model using the first position information and the second position information. 
 
     
     
       11. A traffic event reproduction method comprising:
 receiving first event information including a first event video of a traffic event from a first transmission apparatus, and receiving second event information including a second event video of the traffic event from a second transmission apparatus; 
 generating a first three-dimensional model including an object of a party to the traffic event extracted from the first event video and a second three-dimensional model including an object of the party extracted from the second event video; 
 combining the first three-dimensional model and the second three-dimensional model to generate a combined three-dimensional model in a three-dimensional space including a point corresponding to the occurrence point of the traffic event; and 
 displaying a video of the party viewed from a user-specified viewpoint on a display using the combined three-dimensional model.